Live cell Imaging with fluoview 1000

[Guy Cox]

As I said, I was sticking to the two 'classic' 19th century formulae, from Abbe and Rayleigh.  Rayleigh's formula is for incoherent imaging and fluorescence imaging is incoherent, to all intents and purposes.  But the criterion is arbitrary - if you use full-width half maximum you will get a (slightly) different result.

Abbe's formula assumes coherence but widefield imaging is not truly coherent.  So Michael is quite right.  The question is discussed in Born & Wolf if anyone wants to go there (I don't!).  But the fact remains that the resolution one can obtain in practice, with the best corrected optics, closely matches what Abbe's formula predicts.  For the biologist, going further is pointless.

                                      Guy  



Optical Imaging Techniques in Cell Biology
by Guy Cox    CRC Press / Taylor & Francis
    http://www.guycox.com/optical.htm
______________________________________________
Associate Professor Guy Cox, MA, DPhil(Oxon)
Electron Microscope Unit, Madsen Building F09,
University of Sydney, NSW 2006
______________________________________________
Phone [phone]     Fax [phone]
Mobile [phone]
______________________________________________
     http://www.guycox.net
-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of MODEL, MICHAEL
Sent: Saturday, 21 February 2009 12:47 AM
To: [hidden email]
Subject: Re: Resolution for fluorescence, brightfield and luminescence

The formula for resolution in bright field is not exact. The problem was solved by Hopkins and Barham, Proc Phys Soc B, 63, 737-744, 1950. The effect of condenser on resolution is also shown on the Olympus site, http://www.olympusfluoview.com/java/resolution3d/index.html

Mike  

-----Original Message-----
From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Guy Cox
Sent: Thursday, February 19, 2009 7:49 PM
To: [hidden email]
Subject: Re: Resolution for fluorescence, brightfield and luminescence

In widefield fluorescence the emission wavelength is the only one that matters.  Using the Rayleigh criterion the resolution (r) will therefore be  r = 0.61 x wavelength / NA - though as Julio says this is a somewhat arbitrary criterion.
 
In confocal both wavelengths count, and if the pinhole is infinitely small the Rayleigh resolution r becomes r / sqrt2, so is somewhat better, though most microscopists work with a pinhole size equal to the Airy disk, which will not give any resolution improvement.  However the contribution of the shorter excitation wavelength still means you do slightly better than in widefield (for a simple calculation just assume a wavelength midway between Ex & Em).  
 
In multiphoton fluorescence only the excitation wavelength counts, but the resolution becomes r / sqrt2 without the need for any pinhole - this helps to counteract the loss from the longer wavelength.
 
For more details see Guy Cox & Colin Sheppard, 2004.  Practical limits of resolution in confocal and non-linear microscopy.  Microscopy Research & Technique, 63, 18-22
 
In transmitted light microscopy resolution is determined by diffraction at the sample (which cannot occur in fluorescence) and this is the famous relation given by Abbe:
r = wavelength / 2xNA.  

This assumes that the condenser NA equals or exceeds the objective NA, and so long as this is the case condenser NA doesn't come into it.  With a very small condenser NA (parallel illumination) the resolution is worse:  r = wavelength / NA.  Anything in between will give you resolution in between these extremes.  Thus while condenser NA does affect the result it has much less effect than objective NA.

Unlike Rayleigh, Abbe is a 'hard' criterion - it describes the limiting case which a lens can possibly resolve, since rays from smaller detail will not enter the objective.

I hope this helps,

                                 Guy


 


Optical Imaging Techniques in Cell Biology
by Guy Cox    CRC Press / Taylor & Francis
    http://www.guycox.com/optical.htm
______________________________________________
Associate Professor Guy Cox, MA, DPhil(Oxon) Electron Microscope Unit, Madsen Building F09, University of Sydney, NSW 2006 ______________________________________________
Phone [phone]     Fax [phone]
Mobile [phone]
______________________________________________
     http://www.guycox.net <http://www.guycox.net/>  

 

________________________________

From: Confocal Microscopy List [mailto:[hidden email]] On Behalf Of Julio Vazquez
Sent: Friday, 20 February 2009 4:23 AM
To: [hidden email]
Subject: Re: Resolution for fluorescence, brightfield and luminescence


Hi Monique,  

I won't attempt to give you a definite answer on this, but point you to some literature on the topic:

http://micro.magnet.fsu.edu/primer/anatomy/numaperture.html

http://www.olympusfluoview.com/theory/resolutionintro.html

http://zeiss-campus.magnet.fsu.edu/referencelibrary/pdfs/ZeissConfocalPrinciples.pdf


plus a couple of points:


for fluorescence, I believe the relevant wavelength generally used in the formula would be the average of the peak excitation and peak emission values, although I'm sure the real value would be a more complex one (if your emission as a tail in the longer wavelengths, these will spread your PSF, thus reducing resolution)

The second point that I would stress (and some will probably disagree with this) is that the definition of the minimum resolvable distance is a somewhat arbitrary one. Namely, if you are imaging two diffraction limited spots of similar intensity, their images will be two PSFs, and the minimum resolvable distance will be the distance between the center of the two PSFs so that when you do an intensity profile you see a dip in contrast that you deem is sufficient for you to know with a certain degree of certainty that you have two spots (about 75% of max peak values). Obviously, this depends also on the contrast and noise in your image, and also depends on the fact that you have a priori knowledge that you are looking at two diffraction limited spots. In a real situation, you don't necessarily know that, and therefore you criteria for "certainty" may be a bit different...

Julio.

--
Julio Vazquez, PhD
Director of Scientific Imaging
Fred Hutchinson Cancer Research Center
1100 Fairview Ave. N.,  mailstop DE-512
Seattle, WA 98109-1024

Tel: Office: 206-667-1215/ Lab: 206-667-4205
FAX: [phone]

[hidden email]
http://www.fhcrc.org/science/shared_resources/imaging/


--------------------------------------------------
DISCLAIMER:

This message is confidential, intended only for the named recipient(s) and may contain information that is privileged or exempt from disclosure under applicable law.  If you are not the intended recipient(s), you are notified that the dissemination, distribution or copying of this information is strictly prohibited.  If you received this message in error, please notify the sender then delete this message.



 




On Feb 19, 2009, at 8:10 AM, Vasseur Monique wrote:


        Dear all,

        My question is:
        On the same microscope, same objective, same immersion medium and same
        sample, is there a difference in resolution depending of the microscopy
        method I am using (fluorescence, luminescence or brightfield) since the
        lightpath is not the same?

        Is it correct to consider the following?

        For brightfield:  r = (1.22 * illumination wavelenght)/(NA objective +
        NA condenser)
        For fluorescence and confocal: r = (0.61 * excitation (?) wavelenght) /
        NA objective
        For luminescence: r = (0.61 * emission (?) wavelenght) / NA objective

        Thanks in advance,

        monique



No virus found in this incoming message.
Checked by AVG.
Version: 7.5.552 / Virus Database: 270.10.25/1958 - Release Date: 18/02/2009 8:57 AM



No virus found in this outgoing message.
Checked by AVG.
Version: 7.5.552 / Virus Database: 270.11.1/1961 - Release Date: 19/02/2009 6:45 PM
 

No virus found in this incoming message.
Checked by AVG.
Version: 7.5.552 / Virus Database: 270.11.1/1961 - Release Date: 19/02/2009 6:45 PM
 

No virus found in this outgoing message.
Checked by AVG.
Version: 7.5.552 / Virus Database: 270.11.1/1961 - Release Date: 19/02/2009 6:45 PM
 

[James Pawley]

>The formula for resolution in bright field is not exact. The problem
>was solved by Hopkins and Barham, Proc Phys Soc B, 63, 737-744,
>1950. The effect of condenser on resolution is also shown on the
>Olympus site,
>http://www.olympusfluoview.com/java/resolution3d/index.html
>
>Mike


Hi all,

So I went to the Olympus site and found the very nice JAVA (FLASH?)
interactive resolution demo.

Beautiful!!!

But then I looked a little closer and I am not sure that it is right.
As I remember, Rayleigh criteria is reached when the peak of the Airy
disk of one point overlays the first dark ring of the other AIry
pattern (OK, both peaks overlay the first dark ring of the other
pattern). At his point one sees about 25% contrast (or a dip in the
intensity to about 75% of the peak values on the "Airy Function 3D
Plot").

Now, when you set things on the demo for NA 1.4 and a wavelength of
450 nm, you get what looks like about 25% contrast on the "Airy
Function" at a spacing of 0.2 micrometres which seems about right
although the contrast noted under the Intensity Plot is 31%, not 25%.
Although it is a bit hard to see, I could convince myself that the
peak of one Airy figure seemed to lie over the first dark ring of the
other.

But when you go to NA .75 and a wavelength of 750 nm, at a spacing of
about 0.6 micrometres I get the 25% contrast on both the "Airy
Function" and the readout below the "Intensity Plot", but in the
"Intensity plot" itself, it is very clear that the peak of one Airy
figure is well outside the first dark ring of the other.

More messing led to other situations where the "Contrast" reading
was, say, 98% but the "Intensity Plot" showed nothing close to a
black area between the peaks.

Is this just a math error in the otherwise very beautiful learning
tool or am I missing something?

Maybe this is why I try to stay away from discussions of "resolution"
and try to stick to the Contrast Transfer Function" where it becomes
clear that features that are twice or even 3x the "resolution" in
size are represented in the image at lower contrast than are larger
objects.

Cheers,

Jim P.
--
               **********************************************
Prof. James B. Pawley,                          Ph.  608-263-3147
Room 223, Zoology Research Building,              
FAX  [phone]
1117 Johnson Ave., Madison, WI, 53706  
[hidden email]
3D Microscopy of Living Cells Course, June 13-25, 2009, UBC, Vancouver Canada
Info: http://www.3dcourse.ubc.ca/             Applications due by March 15, 2009
               "If it ain't diffraction, it must be statistics." Anon.